DOCSLIB.ORG

CS675: Convex and Combinatorial Optimization Fall 2019 Introduction to Matroid Theory

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
CS675: Convex and Combinatorial Optimization Fall 2019 Introduction to Matroid Theory CS675: Convex and Combinatorial Optimization Fall 2019 Introduction to Matroid Theory Instructor: Shaddin Dughmi Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets Objective: often “linear”, referred to as modular Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... 1/30 Objective: often “linear”, referred to as modular Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets 1/30 Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets Objective: often “linear”, referred to as modular 1/30 Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets Objective: often “linear”, referred to as modular Analogues of concave and convex: submodular and supermodular (in no particular order!) 1/30 Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets Objective: often “linear”, referred to as modular Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems 1/30 Outline 1 Matroids and The Greedy Algorithm 2 Basic Terminology and Properties 3 The Matroid Polytope 4 Matroid Intersection Maximum Weight Forest Problem Given an undirected graph G = (V; E), and weights we 2 R on edges e, find a maximum weight acyclic subgraph (aka forest) of G. Slight generalization of minimum weight spanning tree We use n and m to denote jV j and jEj, respectively. Matroids and The Greedy Algorithm 2/30 Theorem The greedy algorithm outputs a maximum-weight forest. The Greedy Algorithm 1 B ; 2 Sort non-negative weight edges in decreasing order of weight e1; : : : ; em, with w1 ≥ w2 ≥ ::: ≥ wm ≥ 0 3 For i = 1 to m: S if B feig is acyclic, add ei to B. Matroids and The Greedy Algorithm 3/30 The Greedy Algorithm 1 B ; 2 Sort non-negative weight edges in decreasing order of weight e1; : : : ; em, with w1 ≥ w2 ≥ ::: ≥ wm ≥ 0 3 For i = 1 to m: S if B feig is acyclic, add ei to B. Theorem The greedy algorithm outputs a maximum-weight forest. Matroids and The Greedy Algorithm 3/30 Contrapositive: if B cyclic then so is A Inductively: can extend A by adding jBj − jAj elements from B n A Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic Matroids and The Greedy Algorithm 4/30 Contrapositive: if B cyclic then so is A Inductively: can extend A by adding jBj − jAj elements from B n A Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic (1) and (2) are trivial, so let’s prove (3) Matroids and The Greedy Algorithm 4/30 Contrapositive: if B cyclic then so is A Inductively: can extend A by adding jBj − jAj elements from B n A When jBj > jAj, this means #components(B) < #components(A) Can’t be that all e 2 B are “inside” connected components of A Some e 2 B must “go between” connected components of A. Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic Sub-lemma: if C is acyclic, then jCj = n − #components(C). Induction Matroids and The Greedy Algorithm 4/30 Contrapositive: if B cyclic then so is A Inductively: can extend A by adding jBj − jAj elements from B n A Can’t be that all e 2 B are “inside” connected components of A Some e 2 B must “go between” connected components of A. Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic Sub-lemma: if C is acyclic, then jCj = n − #components(C). Induction When jBj > jAj, this means #components(B) < #components(A) Matroids and The Greedy Algorithm 4/30 Contrapositive: if B cyclic then so is A Inductively: can extend A by adding jBj − jAj elements from B n A Some e 2 B must “go between” connected components of A. Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic Sub-lemma: if C is acyclic, then jCj = n − #components(C). Induction When jBj > jAj, this means #components(B) < #components(A) Can’t be that all e 2 B are “inside” connected components of A Matroids and The Greedy Algorithm 4/30 Contrapositive: if B cyclic then so is A Inductively: can extend A by adding jBj − jAj elements from B n A Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic Sub-lemma: if C is acyclic, then jCj = n − #components(C). Induction When jBj > jAj, this means #components(B) < #components(A) Can’t be that all e 2 B are “inside” connected components of A Some e 2 B must “go between” connected components of A. Matroids and The Greedy Algorithm 4/30 Lemma 1 The empty set is acyclic 2 If A is an acyclic set of edges, and B ⊆ A, then B is also acyclic. Contrapositive: if B cyclic then so is A 3 If A; B are acyclic, and jBj > jAj, then there is e 2 B n A such that A S feg is acyclic Inductively: can extend A by adding jBj − jAj elements from B n A Sub-lemma: if C is acyclic, then jCj = n − #components(C). Induction When jBj > jAj, this means #components(B) < #components(A) Can’t be that all e 2 B are “inside” connected components of A Some e 2 B must “go between” connected components of A. Matroids and The Greedy Algorithm 4/30 Assume true for step i − 1, and consider step i S ∗ If ei 62 Bi, then Bi−1 feig is cyclic. Since Bi−1 ⊆ Bi−1, then ∗ ∗ ∗ ei 62 Bi−1 (Property 2). So take Bi = Bi−1. ∗ ∗ If ei 2 Bi and ei 62 Bi−1, build Bi by repeatedly extending Bi using ∗ Bi−1 (property 3) S ∗ Recall that Bi = Bi−1 feig agrees with Bi−1 on e1; : : : ; ei−1. ∗ ∗ S Bi = Bi−1 feig n fekg for some k > i ∗ ∗ Bi has weight no less than Bi−1, so optimal. Proof Going back to proving the algorithm correct.
Load more

Recommended publications
  • Computational Geometry – Problem Session Convex Hull & Line Segment Intersection
    Computational Geometry { Problem Session Convex Hull & Line Segment Intersection LEHRSTUHL FUR¨ ALGORITHMIK · INSTITUT FUR¨ THEORETISCHE INFORMATIK · FAKULTAT¨ FUR¨ INFORMATIK Guido Bruckner¨ 04.05.2018 Guido Bruckner¨ · Computational Geometry { Problem Session Modus Operandi To register for the oral exam we expect you to present an original solution for at least one problem in the exercise session. • this is about working together • don't worry if your idea doesn't work! Guido Bruckner¨ · Computational Geometry { Problem Session Outline Convex Hull Line Segment Intersection Guido Bruckner¨ · Computational Geometry { Problem Session Definition of Convex Hull Def: A region S ⊆ R2 is called convex, when for two points p; q 2 S then line pq 2 S. The convex hull CH(S) of S is the smallest convex region containing S. Guido Bruckner¨ · Computational Geometry { Problem Session Definition of Convex Hull Def: A region S ⊆ R2 is called convex, when for two points p; q 2 S then line pq 2 S. The convex hull CH(S) of S is the smallest convex region containing S. Guido Bruckner¨ · Computational Geometry { Problem Session Definition of Convex Hull Def: A region S ⊆ R2 is called convex, when for two points p; q 2 S then line pq 2 S. The convex hull CH(S) of S is the smallest convex region containing S. In physics: Guido Bruckner¨ · Computational Geometry { Problem Session Definition of Convex Hull Def: A region S ⊆ R2 is called convex, when for two points p; q 2 S then line pq 2 S. The convex hull CH(S) of S is the smallest convex region containing S.
    [Show full text]
  • On the Ekeland Variational Principle with Applications and Detours
    Lectures on The Ekeland Variational Principle with Applications and Detours By D. G. De Figueiredo Tata Institute of Fundamental Research, Bombay 1989 Author D. G. De Figueiredo Departmento de Mathematica Universidade de Brasilia 70.910 – Brasilia-DF BRAZIL c Tata Institute of Fundamental Research, 1989 ISBN 3-540- 51179-2-Springer-Verlag, Berlin, Heidelberg. New York. Tokyo ISBN 0-387- 51179-2-Springer-Verlag, New York. Heidelberg. Berlin. Tokyo No part of this book may be reproduced in any form by print, microfilm or any other means with- out written permission from the Tata Institute of Fundamental Research, Colaba, Bombay 400 005 Printed by INSDOC Regional Centre, Indian Institute of Science Campus, Bangalore 560012 and published by H. Goetze, Springer-Verlag, Heidelberg, West Germany PRINTED IN INDIA Preface Since its appearance in 1972 the variational principle of Ekeland has found many applications in different fields in Analysis. The best refer- ences for those are by Ekeland himself: his survey article [23] and his book with J.-P. Aubin [2]. Not all material presented here appears in those places. Some are scattered around and there lies my motivation in writing these notes. Since they are intended to students I included a lot of related material. Those are the detours. A chapter on Nemyt- skii mappings may sound strange. However I believe it is useful, since their properties so often used are seldom proved. We always say to the students: go and look in Krasnoselskii or Vainberg! I think some of the proofs presented here are more straightforward. There are two chapters on applications to PDE.
    [Show full text]
  • Parity Systems and the Delta-Matroid Intersection Problem
    Parity Systems and the Delta-Matroid Intersection Problem Andr´eBouchet ∗ and Bill Jackson † Submitted: February 16, 1998; Accepted: September 3, 1999. Abstract We consider the problem of determining when two delta-matroids on the same ground-set have a common base. Our approach is to adapt the theory of matchings in 2-polymatroids developed by Lov´asz to a new abstract system, which we call a parity system. Examples of parity systems may be obtained by combining either, two delta- matroids, or two orthogonal 2-polymatroids, on the same ground-sets. We show that many of the results of Lov´aszconcerning ‘double flowers’ and ‘projections’ carry over to parity systems. 1 Introduction: the delta-matroid intersec- tion problem A delta-matroid is a pair (V, ) with a finite set V and a nonempty collection of subsets of V , called theBfeasible sets or bases, satisfying the following axiom:B ∗D´epartement d’informatique, Universit´edu Maine, 72017 Le Mans Cedex, France. [email protected] †Department of Mathematical and Computing Sciences, Goldsmiths’ College, London SE14 6NW, England. [email protected] 1 the electronic journal of combinatorics 7 (2000), #R14 2 1.1 For B1 and B2 in and v1 in B1∆B2, there is v2 in B1∆B2 such that B B1∆ v1, v2 belongs to . { } B Here P ∆Q = (P Q) (Q P ) is the symmetric difference of two subsets P and Q of V . If X\ is a∪ subset\ of V and if we set ∆X = B∆X : B , then we note that (V, ∆X) is a new delta-matroid.B The{ transformation∈ B} (V, ) (V, ∆X) is calledB a twisting.
    [Show full text]
  • CS675: Convex and Combinatorial Optimization Spring 2018 Introduction to Matroid Theory
    CS675: Convex and Combinatorial Optimization Spring 2018 Introduction to Matroid Theory Instructor: Shaddin Dughmi Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets Objective: often “linear”, referred to as modular Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... 1/30 Objective: often “linear”, referred to as modular Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ... Set system: Pair (X ; I) where X is a finite ground set and I ⊆ 2X are the feasible sets 1/30 Analogues of concave and convex: submodular and supermodular (in no particular order!) Today, we will look only at optimizing modular objectives over an extremely prolific family of set systems Related, directly or indirectly, to a large fraction of optimization problems in P Also pops up in submodular/supermodular optimization problems Optimization over Sets Most combinatorial optimization problems can be thought of as choosing the best set from a family of allowable sets Shortest paths Max-weight matching Independent set ..
    [Show full text]
  • 15 BASIC PROPERTIES of CONVEX POLYTOPES Martin Henk, J¨Urgenrichter-Gebert, and G¨Unterm
    15 BASIC PROPERTIES OF CONVEX POLYTOPES Martin Henk, J¨urgenRichter-Gebert, and G¨unterM. Ziegler INTRODUCTION Convex polytopes are fundamental geometric objects that have been investigated since antiquity. The beauty of their theory is nowadays complemented by their im- portance for many other mathematical subjects, ranging from integration theory, algebraic topology, and algebraic geometry to linear and combinatorial optimiza- tion. In this chapter we try to give a short introduction, provide a sketch of \what polytopes look like" and \how they behave," with many explicit examples, and briefly state some main results (where further details are given in subsequent chap- ters of this Handbook). We concentrate on two main topics: • Combinatorial properties: faces (vertices, edges, . , facets) of polytopes and their relations, with special treatments of the classes of low-dimensional poly- topes and of polytopes \with few vertices;" • Geometric properties: volume and surface area, mixed volumes, and quer- massintegrals, including explicit formulas for the cases of the regular simplices, cubes, and cross-polytopes. We refer to Gr¨unbaum [Gr¨u67]for a comprehensive view of polytope theory, and to Ziegler [Zie95] respectively to Gruber [Gru07] and Schneider [Sch14] for detailed treatments of the combinatorial and of the convex geometric aspects of polytope theory. 15.1 COMBINATORIAL STRUCTURE GLOSSARY d V-polytope: The convex hull of a finite set X = fx1; : : : ; xng of points in R , n n X i X P = conv(X) := λix λ1; : : : ; λn ≥ 0; λi = 1 : i=1 i=1 H-polytope: The solution set of a finite system of linear inequalities, d T P = P (A; b) := x 2 R j ai x ≤ bi for 1 ≤ i ≤ m ; with the extra condition that the set of solutions is bounded, that is, such that m×d there is a constant N such that jjxjj ≤ N holds for all x 2 P .
    [Show full text]
  • Chapter 5 Convex Optimization in Function Space 5.1 Foundations of Convex Analysis
    Chapter 5 Convex Optimization in Function Space 5.1 Foundations of Convex Analysis Let V be a vector space over lR and k ¢ k : V ! lR be a norm on V . We recall that (V; k ¢ k) is called a Banach space, if it is complete, i.e., if any Cauchy sequence fvkglN of elements vk 2 V; k 2 lN; converges to an element v 2 V (kvk ¡ vk ! 0 as k ! 1). Examples: Let ­ be a domain in lRd; d 2 lN. Then, the space C(­) of continuous functions on ­ is a Banach space with the norm kukC(­) := sup ju(x)j : x2­ The spaces Lp(­); 1 · p < 1; of (in the Lebesgue sense) p-integrable functions are Banach spaces with the norms Z ³ ´1=p p kukLp(­) := ju(x)j dx : ­ The space L1(­) of essentially bounded functions on ­ is a Banach space with the norm kukL1(­) := ess sup ju(x)j : x2­ The (topologically and algebraically) dual space V ¤ is the space of all bounded linear functionals ¹ : V ! lR. Given ¹ 2 V ¤, for ¹(v) we often write h¹; vi with h¢; ¢i denoting the dual product between V ¤ and V . We note that V ¤ is a Banach space equipped with the norm j h¹; vi j k¹k := sup : v2V nf0g kvk Examples: The dual of C(­) is the space M(­) of Radon measures ¹ with Z h¹; vi := v d¹ ; v 2 C(­) : ­ The dual of L1(­) is the space L1(­). The dual of Lp(­); 1 < p < 1; is the space Lq(­) with q being conjugate to p, i.e., 1=p + 1=q = 1.
    [Show full text]
  • Matroid Intersection
    EECS 495: Combinatorial Optimization Lecture 8 Matroid Intersection Reading: Schrijver, Chapter 41 • colorful spanning forests: for graph G with edges partitioned into color classes fE1;:::;Ekg, colorful spanning forest is Matroid Intersection a forest with edges of different colors. Define M1;M2 on ground set E with: Problem: { I1 = fF ⊂ E : F is acyclicg { I = fF ⊂ E : 8i; jF \ E j ≤ 1g • Given: matroids M1 = (S; I1) and M2 = 2 i (S; I ) on same ground set S 2 Then • Find: max weight (cardinality) common { these are matroids (graphic, parti- independent set J ⊆ I \I 1 2 tion) { common independent sets = color- Applications ful spanning forests Generalizes: Note: In second two examples, (S; I1 \I2) not a matroid, so more general than matroid • max weight independent set of M (take optimization. M = M1 = M2) Claim: Matroid intersection of three ma- troids NP-hard. • matching in bipartite graphs Proof: Reduction from directed Hamilto- For bipartite graph G = ((V1;V2);E), let nian path: given digraph D = (V; E) and Mi = (E; Ii) where vertices s; t, is there a path from s to t that goes through each vertex exactly once. Ii = fJ : 8v 2 Vi; v incident to ≤ one e 2 Jg • M1 graphic matroid of underlying undi- Then rected graph { these are matroids (partition ma- • M2 partition matroid in which F ⊆ E troids) indep if each v has at most one incoming edge in F , except s which has none { common independent sets = match- ings of G • M3 partition :::, except t which has none 1 Intersection is set of vertex-disjoint directed = r1(E) +r2(E)− min(r1(F ) +r2(E nF ) paths with one starting at s and one ending F ⊆E at t, so Hamiltonian path iff max cardinality which is number of vertices minus min intersection has size n − 1.
    [Show full text]
  • The Krein–Milman Theorem a Project in Functional Analysis
    The Krein{Milman Theorem A Project in Functional Analysis Samuel Pettersson November 29, 2016 2. Extreme points 3. The Krein{Milman theorem 4. An application Outline 1. An informal example 3. The Krein{Milman theorem 4. An application Outline 1. An informal example 2. Extreme points 4. An application Outline 1. An informal example 2. Extreme points 3. The Krein{Milman theorem Outline 1. An informal example 2. Extreme points 3. The Krein{Milman theorem 4. An application Outline 1. An informal example 2. Extreme points 3. The Krein{Milman theorem 4. An application Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". kxk1≤ 1 Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". kxk1≤ 1 Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". kxk1≤ 1 kxk2≤ 1 Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". kxk1≤ 1 kxk2≤ 1 Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". kxk1≤ 1 kxk2≤ 1 kxk1≤ 1 Convex sets and their \corners" Observation Some convex sets are the convex hulls of their \corners". kxk1≤ 1 kxk2≤ 1 kxk1≤ 1 x1; x2 ≥ 0 Convex sets and their \corners" Observation Some convex sets are not the convex hulls of their \corners". Convex sets and their \corners" Observation Some convex sets are not the convex hulls of their \corners".
    [Show full text]
  • Packing of Arborescences with Matroid Constraints Via Matroid Intersection
    Mathematical Programming (2020) 181:85–117 https://doi.org/10.1007/s10107-019-01377-0 FULL LENGTH PAPER Series A Packing of arborescences with matroid constraints via matroid intersection Csaba Király1 · Zoltán Szigeti2 · Shin-ichi Tanigawa3 Received: 29 May 2018 / Accepted: 8 February 2019 / Published online: 2 April 2019 © The Author(s) 2019 Abstract Edmonds’ arborescence packing theorem characterizes directed graphs that have arc- disjoint spanning arborescences in terms of connectivity. Later he also observed a characterization in terms of matroid intersection. Since these fundamental results, intensive research has been done for understanding and extending these results. In this paper we shall extend the second characterization to the setting of reachability-based packing of arborescences. The reachability-based packing problem was introduced by Cs. Király as a common generalization of two different extensions of the spanning arborescence packing problem, one is due to Kamiyama, Katoh, and Takizawa, and the other is due to Durand de Gevigney, Nguyen, and Szigeti. Our new characterization of the arc sets of reachability-based packing in terms of matroid intersection gives an efficient algorithm for the minimum weight reachability-based packing problem, and it also enables us to unify further arborescence packing theorems and Edmonds’ matroid intersection theorem. For the proof, we also show how a new class of matroids can be defined by extending an earlier construction of matroids from intersecting submodular functions to bi-set functions based on an idea of Frank. Mathematics Subject Classification 05C05 · 05C20 · 05C70 · 05B35 · 05C22 B Csaba Király [email protected] Zoltán Szigeti [email protected] Shin-ichi Tanigawa [email protected] 1 Department of Operations Research, ELTE Eötvös Loránd University and the MTA-ELTE Egerváry Research Group on Combinatorial Optimization, Budapest Pázmány Péter sétány 1/C, 1117, Hungary 2 Univ.
    [Show full text]
  • Convexity in Oriented Matroids
    View metadata, citation and similar papers at core.ac.uk brought to you by CORE provided by Elsevier - Publisher Connector JOURNAL OF COMBINATORIAL THEORY, Series B 29, 231-243 (1980) Convexity in Oriented Matroids MICHEL LAS VERGNAS C.N.R.S., Universite’ Pierre et Marie Curie, U.E.R. 48, 4 place Jussieu, 75005 Paris, France Communicated by the Editors Received June 11, 1974 We generalize to oriented matroids classical notions of Convexity Theory: faces of convex polytopes, convex hull, etc., and prove some basic properties. We relate the number of acyclic orientations of an orientable matroid to an evaluation of its Tutte polynomial. The structure of oriented matroids (oriented combinatorial geometries) [ 1 ] retains many properties of vector spaces over ordered fields. In the present paper we consider classical notions such as those of faces of convex polytopes, convex hull, etc., and show that usual definitions can be extended to finite oriented matroids in a natural way. We prove some basic properties: lattice structure with chain property of the set of faces of an acyclic oriented matroid, lower bounds for the numbers of vertices and facets and charac- terization of extremal cases, analogues of the Krein-Milman Theorem. In Section 3 we relate the number of acyclic orientations of an orientable matroid A4 to the evaluation t(M; 2,0) of its Tutte polynomial. The present paper is a sequel to [ 11. Its content constituted originally Sections 6, 7 and 8 of “MatroYdes orientables” (preprint, April 1974) [8]. Basic notions on oriented matroids are given in [ 11. For completeness we recall the main definitions [ 1, Theorems 2.1 and 2.21: All considered matroids are on finite sets.
    [Show full text]
  • The Convex Hull
    Chapter 6 The convex hull The exp ected value function of an integer recourse program is in general non- convex. The lack of convexity precludes the use of many results that are in- disp ensable for eciently solving these mo dels. It is well-known that, given convexity, every lo cal minimum is a global minimum. Moreover, the duality theory for convex optimization problems provides a wealth of results, resulting in the fact that such problems are well-solved, at least in theory (see Grotschel et al. [14]). For example, the entire theory and all algorithms to solve continu- ous recourse programs are based on the convexity of these problems. These considerations justify our interest in convex approximations of the exp ected value function of integer recourse mo dels. Obviously, if the exp ected value function is replaced by such an approximation, the resulting mo del is a convex minimization problem, since the rst-stage ob jective function is linear and the constraints are linear equalities and non-negativities. In this and the following chapter we deal with convex approximations of the integer exp ected value function Q, in case the recourse is simple and the technology matrix T is xed. For easy reference, we rep eat its de nition: n 1 ; Q(x) = E v (x; ! ); x 2 R ! + + + v (x; ! ) = inf fq y + q y : y p(! ) T x; y p(! ) T x; m m + 2 2 y 2 Z ; y 2 Z g; + + + where (q ; q ) and T are xed vectors/matrices of compatible dimensions, + q 0, q 0, and p(! ) is a random vector.
    [Show full text]
  • Lecture 12 1 Matroid Intersection Polytope
    18.438 Advanced Combinatorial Optimization October 27, 2009 Lecture 12 Lecturer: Michel X. Goemans Scribe: Chung Chan (See Sections 41.4 and 42.1 in Combinatorial Optimization by A. Schrijver) 1 Matroid Intersection Polytope For matroids M1 = (S, I1) and M2 = (S, I2), the intersection polytope is the convex hull of the incidence vectors of the common independent sets: Conv{χ(I),I ∈ I1 ∩ I2}. (1) S Theorem 1 Let r1, r2 : 2 7→ Z+ be the rank functions of the matroids M1 = (S, I1) and M2 = (S, I2) respectively. Then, the intersection polytope (1) is equivalent to the set of S x ∈ R defined by the following totally dual integral (TDI) system, X x(u) := xe ≤ ri(U) ∀U ⊆ S, i = 1, 2 (2a) e∈U x ≥ 0 (2b) Proof: It suffices to show that (2) is TDI because then all extreme points are integral, and so the set of vertices are the set of incidence vectors of the common independent sets in both matroids. To prove TDI, consider the following linear programming duality for some weight vector S w ∈ Z , T X X max w x =minimize r1(U)y1(U) + r2(U)y2(U) (3a) x≥0:x(U)≤r (U),i=1,2 i U⊆S U⊆S X subject to [y1(U) + y2(U)] ≥ we ∀e ∈ S (3b) U3e with y1, y2 ≥ 0 Observe that we can assume that w ≥ 0 since any negative entry can be deleted as it does not affect feasibility in the dual (3). ∗ ∗ Claim 2 There exists an optimal solution (y1, y2) to the dual problem in (3) that satisfies, ∗ y1(U) = 0 ∀U/∈ C1 (4a) ∗ y2(U) = 0 ∀U/∈ C2 (4b) S for some chains C1, C2 ⊆ 2 , where a chain F is such that, for all A, B ∈ F, either A ⊆ B or B ⊆ A.
    [Show full text]